My first Dara App: Wine Classification Problem

---

A. Screenshots

Home page

EDA page

Classification page

B. Installation and Setup

1. Poetry

• Install curl -sSL https://install.python-poetry.org | python3 -

• Add export PATH="$HOME/.local/bin:$PATH" to .zshrc (source)

• Use poetry --version

2. Create new app and install dara

• Create new app

  • In PyCharm with new poetry env
  • manually using mkdir and poetry init

• Run poetry install

• Install poetry add dara-core --extras all

C. Creating a basic app tutorial

1. Hello World

• Add base code to main.py

• Run poetry run dara start

• Visit http://0.0.0.0:8000/

2. Add new pages

• Install dependencies to build a page that cna handle data: poetry add scikit-learn

• Create a pages subdir and add new pages in python there

• Add new pages to the main config using config.add()

3.Future Components

• Choose model from list of model e.g., RF, GBMs, SVM, ...
• More model validation metrics (ROC curve, precision/recall trade-off)
• More EDA (similar to pandas-profiling)

Contact

• Email address: adam[dot]jaamour[at]causalens[dot]com